The present invention belongs to an optical head used in an optical disk device, and more particularly relates to a technique for enhancing a performance in detection of a position controlling signal for an optical spot thereof.
Conventional techniques on methods for controlling a focal point position in an optical disk device are described in, for example, “Fundamentals and Applications of Optical Disk Storage”, Y. Tsunoda, 1995, 1st edition (Korona Corp., Tokyo), pp. 79-83. According to this literature, there are the following methods: Foucault method (Knife edge method), an astigmatic method, a beam size detection method, an image rotating method, and so on. From criteria such as simplicity of an optical system required, the ease with which the adjustment can be made, and the ease with which combination with a tracking detection can be achieved, the most prevailing method, at the present stage, is the astigmatic method. In the astigmatic method, however, there existed a problem that, when an optical spot crosses a track on the surface of a storage film, a disturbance is apt to occur in a focus error signal in association with a decentering of an optical disk. This disturbance is likely to occur especially when astigmatism takes place in a focused spot or the optical spot is shifted on an optical detector. Examples of methods for reducing the disturbance are disclosed as follows: A method of reducing the disturbance by blocking light out of a central portion of a detected light beam is disclosed in JP-A-6-162527 and JP-A-6-309687, a method of reducing it by adjusting rotation of an objective lens is disclosed in JP-B-5-68774, and a method of reducing it by means of an operation between a light with astigmatism and a light without astigmatism in a detected system is disclosed in JP-A-5-197980. None of them, however, is a fundamental method for solving the above-described problem. Thus, at the present stage, the reducing effect obtained is not necessarily enough.
In particular, in a land-groove type optical disk employed in a DVD-RAM planned to be brought into a commercial stage soon, the disturbance occurs quite outstandingly. The reason is as follows: In the land-groove type optical disk, a width of a guiding groove (groove) is substantially equal to a width of a portion of a guiding inter-groove (land), and information is stored on the both sides thereof. On account of this, a pitch of the guiding groove itself, when compared with an optical spot, is formed to be larger than in conventional optical disks. This extraordinarily intensifies a tracking error signal according to a push-pull method described later, thus causing the disturbance to occur quite outstandingly. This condition, accordingly, brings about a situation that, in an optical head for the DVD-RAM, it can not be helped employing the Foucault method or the beam size detection method the configuration or the adjustment of which is complicated.
Conventional techniques on methods for controlling a tracking in an optical disk device are similarly described in, for example, the above-cited “Fundamentals and Applications of Optical Disk Storage”, Y. Tsunoda, 1995, 1st edition (Korona Corp., Tokyo), pp. 83-92. According to this literature, there are methods such as a three-spots method and a diffracted light differential method (push-pull method). Judging from criteria such as simplicity of an optical system required, the ease with which the adjustment can be made, and a resistance to the disturbance, the three-spots method is mainly employed in a read only type optical disk such as a compact disk (CD). Meanwhile, the push-pull method is mainly employed in the case of a magneto-optical disk or the DVD-RAM which needs a high laser emission power at the time of the recording. At this time, there can be considered another way in which, exchanging the roles with each other, the push-pull method is employed toward the CD and the three-spots method is employed toward writable optical disks. However, there exist circumstances which make such an employment impossible.
In performing a CD pick up, in order to cause a focused spot to follow a decentering of the optical disk for the necessity of low price, the objective lens is moved by only being mounted on a lens actuator. Then, if the push-pull method is employed, it turns out that a detected light beam moves on the optical detector. This phenomenon appears as an off-set. Also, at a pit depth of λ/(4n) (λ: light wavelength, n: substrate refractive index) at which a signal amplitude becomes largest in the reproduction-only type optical disk, there are the following problems: Of diffracted light by means of a periodic structure of train of pits in the radial direction, 0th order light becomes smaller. In addition to this, even when the focused spot is off-track, no unbalance occurs in interference intensity between the 0th order light and ±1st order diffracted lights. This makes it impossible to obtain the tracking error signal.
Meanwhile, in the storage-able optical disks, especially in the magneto-optical disk, compensation for the decentering of the optical disk is usually performed by an actuator called a coarse actuator. The coarse actuator mounts the optical head or only a portion of the objective lens and an objective lens actuator so as to allow the optical spot to come near to a proximity of a track to be objected. Namely, the magneto-optical disk is constituted in such a manner that, of a tracking error, the low frequency components are compensated by the coarse actuator and the high frequency components are compensated by the objective lens actuator, thereby enhancing a reliability needed for the storage operation. Accordingly, an amount of movement by the objective lens actuator is lower than in the read only type optical disk such as the CD. This makes it possible to employ the push-pull method which has higher light utilization efficiency than the three-spots method does.
Also, if the three-spots method is employed toward the storage-able optical disks, as described on page 127 of “Technical Digest of Symposium on Optical Memory '86”, there take place the following problems: First, in an optical disk such as the DVD-RAM, i.e. the type of optical disk that performs the storage by means of a variation in reflectance of a storage mark, at the time of the storage operation, there arises a difference in the amount of light between a preceding sub-spot and a subsequent sub-spot. This causes an off-set to occur in the tracking error signal. Also, in the case of the magneto-optical disk, there exists a feedback light back to a semiconductor laser. On account of this, a tilt of the disk unbalances a condition of stray-lights interference on the both sides of sub-spots. This also causes an off-set to occur. Moreover, as described already, the land-groove type optical disk is employed in the DVD-RAM. This circumstance can also be mentioned as a reason for making it impossible to employ the three-spots method toward the DVD-RAM. Namely, in the land-groove type optical disk, a width of the land portion is originally made equal to that of the groove portion in order to make an amount of reflected light of the land portion equal to that of the groove portion. This necessarily results in a fact that, even when an optical spot is off-track, the amount of light scarcely varies, thus making it impossible to obtain a tracking error signal according to the three-spots method. Accordingly, it can not be helped employing the push-pull method in the DVD-RAM. However, unlike the case of the magneto-optical disk, it is required to lower the price of the DVD-RAM down to a price close to the price of the CD. Consequently, it becomes absolutely necessary to reduce the off-set in a tracking error signal which accompanies the movement of the objective lens according to the push-pull method.
A conventional technique for solving the above-mentioned problem in the DVD-RAM is described in, for example, “National Technical Report”, Vol. 40, No. 6, (1994), pp. 771-778. Here, the optical disk device is constituted as follows: The objective lens, a λ/4 plate, and a polarizing diffraction grating are integrally mounted on an objective lens actuator. Moreover, the polarizing diffraction grating is constituted so that interference regions, in which, of diffracted light by mean of the disk, +1st order diffracted light and −1st order diffracted light each interfere with 0th order light, are diffracted with a different diffraction angle, respectively. This constitution makes it possible to separate, on the optical detector, the interference region between the +1st order diffracted light and the 0th order light from the interference region between the −1st order diffracted light and the 0th order light. From this, the above-mentioned literature shows the following: If a dual-divided optical detector is constituted so that, when the objective lens is moved, the lights do not stray out of the optical detector, it becomes possible to eliminate the off-set caused by the phenomenon that the optical spots move on the optical detector.
Also, employing the polarizing diffraction grating as a diffraction grating makes the following possible: When a light heading for the disk passes through the polarizing diffraction grating, the diffraction efficiency is made substantially equal to zero, and when a reflected light from the disk passes through the polarizing diffraction grating again, the diffraction efficiency is caused to become an appropriate value. Meanwhile, in the case of a non-polarizing ordinary diffraction grating, it diffracts the light heading for the disk, too, thus making it impossible to avoid a loss of the amount of light. Employing the polarizing diffraction grating in this way allows only the necessary diffraction of the reflected light to occur, thus making it possible to prevent the loss of the amount of light.
However, in this conventional example, the objective lens, the λ/4 plate, and the polarizing diffraction grating are integrally mounted on the objective lens actuator. This constitution results in a problem that a movable portion of the actuator becomes heavy, thus restricting a response speed of the actuator down to a low level. Since optical disks are being improved in the storage density and at the same time are becoming faster in the transfer rate year by year, the above-described conventional example is not able to meet the trend of even further speeding-up in the near future.
Another method, which, with no other optical component except the objective lens mounted on the objective lens actuator, makes it possible to eliminate the tracking error signal off-set which accompanies the movement of the objective lens according to the push-pull method, is disclosed in the above-described “Technical Digest of Symposium on Optical Memory '86”, pp. 127-132. This method is referred to as a differential push-pull method. In the method, the three-spots method is employed, and respective tracking error signals according to the push-pull method are subtracted on a main-spot and two sub-spots, thereby eliminating the tracking error signal off-set which accompanies the movement of the objective lens. Namely, in the method, the sub-spots are located in such a manner that they are shifted on the both sides of the main spot by one-half of a period of the guiding groove, thereby simultaneously detecting a light beam in which variations in interference intensity distribution of a reflected light beam reflected from the disk in association with an off-track are inverted, and thus generating opposite phase tracking error signals which contain the off-set in the same phase. Then, these opposite phase tracking error signals are subtracted, thereby allowing only the off-set to be cancelled. According to this conventional example, the ratio of the gain to amplify the signal by the main spot to the gain of the signal by the sub-spot is chosen so as to compensate the intensity unbalance caused by diffraction efficiency characteristics of the diffraction grating to generate sub-spots. The use of this conventional example, with no other optical component except the objective lens mounted on the objective lens actuator, basically makes it possible to eliminate the tracking error signal off-set which accompanies the movement of the objective lens according to the push-pull method. In the present conventional example, however, no countermeasure is taken against the mixture of the disturbance into the focus error signal when a focused spot crosses the guiding groove in the astigmatic focus error detecting method described earlier. Also, as described in the present conventional example, when one of the sub-spots lies in a post-stored track and the other lies in a pre-stored track, the effect of reducing the off-set is not enough. Further, although not described in the present conventional example, when a total amount of reflected light on the guiding grooves differs from a total amount of reflected light on the guiding inter-grooves, the off-set also remains in the present conventional example. This situation arises when a width of the guiding groove is not equal to that of the guiding inter-groove. However, in the case of the DVD-RAM employing the land-groove type optical disk in which the width of the guiding groove is equal to that of the guiding inter-groove, such a situation also arises if the main-spot lies in the post-stored track and the two sub-spots lies in the pre-stored track or in the case opposite thereto. Still further, in the present conventional example, there exist the plurality of optical spots. This brings about a disadvantage in the light utilization efficiency at the time of the storage.
Moreover, the gains to amplify the signals by main spot and sub-spots chosen in this conventional method is insufficient to cancel the effect completely. The reason is as follows. As described later, when a width of the guiding groove does not substantially equal to half of the pitch of guiding grooves, the reflectance of the light when the focused spot is at the guiding groove is different from that when the focused spot is at the inter-groove. It is also necessary to compensate this unbalance of reflectance for perfect offset cancellation. For the higher the density of the optical disk, the allowance of the offset is the severer. Therefore this insufficient cancellation must be a problem, recently.
Still more, in this conventional example, the optical disk was not a land-groove type optical disk. Therefore, there is no cross-talk from stored information signal, because the sub-spots on the optical disk are not on the information track at readout process. In the case of land-groove type optical disk such as DVD-RAM, however, the sub-spots is also on the information track of reading out from the disk. This results in disturbance to the tracking error signal.
Another method, which cancels the disturbance in the focus error signal of astigmatic method, is disclosed in the JP-A-4-168631. Also in this method, the main spot and sub-spots by a diffraction grating is positioned onto the optical disk at the distances of the half of the pitch of guiding grooves in the radial direction of the disk. The reflected beams from these focused spots pass through a cylindrical lens, then detected by three quadratic photo-detectors, respectively. From the output signal of these photo detectors, three focus error signals are obtained by calculation in the electric circuit. These focus error signals are amplified with gains which are proportional to the reciprocals of the light intensity of each focused spot on the optical disk, which are not proportional to the reciprocal of the reflected light intensity. Then summation of these amplified focus error signals is calculated in the electric circuit. Employing this method, the extra disturbance to the conventional focus error signal by aberrations or miss-alignment of the optical components or photo detector can be eliminated. The optimum gains for disturbance cancellation for this method is different from those for differential push-pull method as mentioned. However, in this method, no tracking method is disclosed. Further more, if the differential push-pull method described in the conventional literature itself is employed with this focusing error detection method, it is necessary to set the gains to amplify the signal by each reflected light beam equal between in the focus error signal and in the tracking error signal, namely proportional to the reciprocals of the light intensity of incident focused spots on to the optical disk. It results in the insufficient cancellation of the offset of tracking error signal as mentioned.